Spreadsheet

You can use this MCP server to read from and write to Google Sheets by issuing Model Context Protocol actions through an MCP client. It enables your AI assistants to reference spreadsheets, fetch data, and perform structured updates in a safe, programmatic way.

How to use

You interact with the server through an MCP client. Use the available read tools to inspect spreadsheets, verify structures, and retrieve data. When you are ready to modify sheets, you can employ the write tools to create spreadsheets, add sheets, set headers, and append rows. For formatting tasks, you can apply styling and validation rules to keep data clean and consistent.

How to install

Prerequisites: you need Node.js installed on your machine. Ensure you have a working internet connection to install dependencies and authenticate with Google.

# 1. Clone the project directory if not already present
# (use your preferred method to obtain the server source)

# 2. Install dependencies
cd spreadsheet-mcp
npm install

# 3. Build the project
npm run build

# 4. Start the server (or follow the runtime instructions below after building)
npm start
```,

Configuration and starting the MCP server

This server uses a credential store for Google authentication and is designed to be launched as an MCP endpoint. During the first run, your browser will open to complete the authentication flow with your Google account.

The MCP connection is configured in Cursor’s settings. You add a local runtime entry that points to the built JavaScript file.

{
  "mcpServers": {
    "spreadsheet": {
      "command": "node",
      "args": ["~/spreadsheet-mcp/dist/index.js"]
    }
  }
}
```,

What you can do with the available tools

Read operations: get_sheet_metadata, get_sheet_data, get_column_values, get_sheet_structure, search_in_sheet, check_sheet_errors.

Write operations: create_spreadsheet, create_sheet, set_headers, append_rows.

Formatting operations: format_sheet, set_data_validation.

Security and safety

If you enable writing tools, you can use the dry_run option to preview changes without applying them. This helps you validate actions before making any modifications.

Development

If you are actively developing, you can build and run in development mode using the project’s scripts. Typical commands are build followed by start or a dedicated dev command if provided.

Available tools

get_sheet_metadata

Retrieve the list of sheets, column names, and row counts.

get_sheet_data

Fetch data from a specified cell range within a sheet.

get_column_values

Obtain all values from a specific column.

get_sheet_structure

Infer headers, data types, and sample data for a sheet.

search_in_sheet

Search for a specific string within a sheet.

check_sheet_errors

Detect cell errors such as #REF!, #VALUE!, and others.

create_spreadsheet

Create a new Google Spreadsheet.

create_sheet

Add a new sheet to an existing spreadsheet.

set_headers

Set or update the header row for a sheet.

append_rows

Append data rows to a sheet.

format_sheet

Apply header styling, column widths, and alternating row colors.

set_data_validation

Configure dropdowns, numeric ranges, and other input rules.